I won’t lie, I like Microsoft.  I think Windows is the best all-round family of operating systems available.  And I use Internet Explorer as my default browser, even though I have Firefox and Chrome installed.

But I also run a Linux server.  It is a modest beast.  It has a Sempron processor, three hard disks around 200-500GB each, and about 1GB of RAM.  It doesn’t need much, even though it acts as a mail server, a web server, and a DNS and network file server for my home LAN.  It runs Fedora 11, which is actually quite nice.

I started using Linux with no experience, and with the help of some patience, a few good Internet resources, and good old intuition, I pretty much know what I’m doing.

About a week ago, my Internet connection started to die sporadically, at unpredictable intervals, for no apparent reason.  I traced the lack of connectivity down to the DNS server not responding to requests, and this led me to realise that the Linux box would not respond to any kind of request at all: SSH, HTTP, or even ping.

So imagine my surprise, when after a lot of investigation (and I really do mean a LOT of investigation) it turned out to be Samba, the service which handles network file shares.

I had recently updated Samba from v3.3 to v3.4 using yum, assuming it to be a bug-free release which wouldn’t break anything.  How wrong I was.  From v3.4, Samba has been changed to use a different authentication mechanism by default.  It used to use smbpasswd up to v3.3, but now uses tdbsam.

Ok, no problem really – until the upgrade from v3.3 to v3.4 changes your smb.conf without leaving an smb.conf.backup or something similar sitting next to it.  So without knowing it, your Samba installation now uses tdbsam.  Now, the problem with this is that there is a known bug whereby Windows clients become unable to authenticate and connect.

It was disappointing to have to find this out the hard way – by which I mean trawling around the web.  But still, it should be easy enough to edit smb.conf and change the authentication back to smbpasswd, right?  Wrong, because it seems the upgrade deleted the file containing the smbpasswd credentials (/var/lib/samba/private/smbpasswd).

Redhat’s Bugzilla entry explains how to convert smbpasswd account credentials to tdbsam format, so presumably you can do the reverse.  Well, technically you can – but only if the tdbsam credentials file (/var/lib/samba/private/passdb.tdb) isn’t empty!  That’s right, the upgrade failed (or didn’t even try) to convert the smbpasswd account credentials to the tdb format.

The solution is to run smbpasswd -a username for every Samba account you lost.  I’m glad I only had two.

This is why I think Linux is a long long way from being ready to take over the world.  I don’t think the Linux community appreciates how much help the average desktop user needs when they encounter problems.  This is demonstrated by the fact that Linux gives you no help at all; even the noddy ‘How do I…’ help topics in Windows destroy the assistance Linux offers.

From the very start, Linux makes it very clear you’re on your own.

While I admit that the change to Samba was publicised in the Samba release notes, I simply do not have the time to read through the release notes of, and perform impact analysis upon, every update I install.  I want to be confident that each update is the best thing for my system – much like Windows Update.

For now, I will stick with the majority of users, and keep current with Windows updates for the advantage of not having to enter the root user’s password whenever I want to do something.

There are several RAID configurations.  RAID 0 (striping) utilises two or more identical hard drives, and splits each file equally across them, yielding much faster disk performance (limited by the throughput of the disk I/O controller) as the disks read and write at the same time.  The total available storage is the sum of the individual drives’ capacity.  RAID 1 (mirrored) could be considered the opposite of striping, in that each file is written to all disks.  This effectively provides realtime backup since the content of all drives in the array is identical, with no performance cost.

I’ve only ever used RAID 0.  For a home PC, RAID 1 doesn’t offer any real benefits that cannot be achieved with a half-decent backup policy, but effectively doubles the cost of disk space if you use two disks.

The only down-side to RAID that I had found was that drivers need to be provided during the installation of Windows.  Linux is slightly better in this respect because it has native drivers for the most popular RAID controllers, usually providing basic functionality.  Windows, however, asks that you provide the drivers on installation, presumably so that you benefit from all the features that the RAID controller offers.  With Windows XP you were required to hit F6 during the installation, and provide the drivers on a floppy disk (which was a real pain, since floppy drive usage died out shortly after the discovery of fire, and RAID drivers often came on a CD anyway).  With Vista and Windows 7 the installer is slightly more intelligent.  It will recognise the presence of a RAID array, but it will still require drivers to be provided.

This is all fine; once you’re familiar with the quirks of the Windows installer it’s plain sailing.

But RAID 0 has its own special drawback, especially when you’re not expecting it.  If you take a disk from an old RAID 0 array, install it as the main disk in a PC and try to install Windows 7, you’ll have a nasty surprise.  Usually the Windows installer will acknowledge the presence of a drive, even if it can’t do anything with it.  But not with an ex-mirrorred array disk.  The Windows installer doesn’t even recognise the disk as being present.

Presumably this is related to the fact that the RAID controller had written its own form of partition information to the disk; while the Windows installer can partition and format a raw disk, or one that is already partitioned, it cannot do anything with a disk whose partition information is present but apparently corrupt.

And the icing on the cake?  You can’t repartition the disk without it being partitioned..!

Yesterday I made the decision (read: mistake) to update my Netgear DG834G router (hardware v4, firmware v5.01.09) to firmware v5.01.14 – and, as is the way with these things, it brought trouble.  After the upgrade I couldn’t reach www.nikrivers.com from the LAN side of the router.

The problem is caused by the way the router handles traffic coming from an internal IP address and destined for the WAN (i.e. external) IP address.  In this situation it requires that the router first transfers the traffic from the internal network to the external network, and then immediately passes it back whilst applying any firewall or routing rules that are relevant to incoming external traffic.

This behaviour is called ‘NAT loopback’, and it seems the vast majority of routers built for the home market have this ability turned off, or do not have the ability at all.  It can be a big problem if you host a website and wish to access that same website using its domain name.  The domain name will resolve to the WAN IP address of your router, and any traffic headed there (such as an HTTP GET request on port 80) from the internal network will be ignored by the router.

There are a few ways to solve this, but none of them are ideal.

1. Use the server name instead of the domain name to access your website
2. Modify the list of known network hosts on each client to point your domain name straight to the server in question.
3. Run your own DNS server, using a view to return the server’s local IP address to requests for your domain name originating from your network.

Of course, the situation gets more complex if you’re also using your router to send TCP traffic on port 80 to your webserver and UDP traffic on port 8668 to a game server.

The solution is to get NAT loopback working on your router.  With some routers, such as the Touchspeed 535 as provided by Be Broadband, this feature can be enabled using a simple CLI command.  For other routers, such as the Netgear DG834G, it’s not quite so easy.

For the purposes of this post I’ll assume the internal network is on the 192.168.0.x range, the router is 192.168.0.254, and the web server is 192.168.0.1.  You will need to modify these IP addresses according to your own network setup.

The first thing to do is to enable debug mode on the router.  Simply go to http://192.168.0.254/setup.cgi?todo=debug and you’ll be rewarded with an appropriate message, “Debug Enable!”.  Nice.

Now connect to the router with ‘telnet 192.168.0.254′ to gain access to the router’s cut-down installation of Linux.  All that is required is to add one additional entry to the router’s iptables (which is a standard Linux feature; Google it or more info).  Type the following, amending any IP addresses according to the network setup:

iptables -t nat -A POSTROUTING -d 192.168.0.1 -s 192.168.0.0/24 -p tcp –dport 80 -j SNAT –to 192.168.0.254

This adds a rule to the POSTROUTING chain on the nat table which applies to all TCP traffic on port 80 (HTTP) coming from the private network and headed to the router.  The rule redirects the traffic to the server, and then processing jumps to the SNAT chain.

If the server is more than just a simple web server, such as an NTP server or mail server as well, the above step needs to be performed (changing the -p and –dport parameters accordingly) for each port and protocol combination you require.  Alternatively, those parameters could be omitted altogether, which will allow all traffic types on all ports through:

iptables -t nat -A POSTROUTING -d 192.168.0.1 -s 192.168.0.0/24 -j SNAT –to 192.168.0.254

If you do this, I recommend you run a firewall on your server, with only the appropriate ports opened.

There is more information in section 10 of Rusty Russell’s Linux 2.4 NAT Howto.

Unfortunately, the iptables change isn’t retained when the router restarts, so it is necessary to go through the process every time – which is a pain in the backside.  Fortunately, however, the Netgear support website has a download link for previous firmware versions, so I downgraded my router back to firmware v5.01.09 and everything worked fine again–including NAT loopback–with no iptables hack required.

A robust solution with DNS

Simply put, proper DNS is the best way to get around a router’s lack of/poorly implemented NAT loopback.

If you have the resources to host a website then you most likely also have the resources to host a DNS server for your internal network.  Simply create an ACL list describing all the clients on your internal network (probably as simple as specifying the CIDR block for your network, maybe something like 192.168.1.0/24).  Then create a view whose clients match that ACL, and define that view as a master DNS server for your website domain.  You then need to create a zone file for that domain – but instead of using an external IP for your webserver, use its internal IP.

All requests for other domains will be routed to the DNS forwarders, but requests for your webserver’s domain will be handled locally, and internal IP addresses will be returned.

The benefit is that you avoid traversing your gateway router to simply come back inside your network.  It doesn’t make sense that you rely on your gateway router to access a website within your own network.  In addition, the firewall on your router can be hardened to a much greater degree: for example, you needn’t leave FTP ports open on your router if you’re only connecting locally.  Or, to put it another way, you’re likely going to want to give yourself more access to your server than you want to give to the outside world; configuring a router’s firewall rules for this kind of conditional logic is simply asking for trouble.

There were no serious problems, but these failures started becoming regular.  I realised it was because I was running Fedora 9, and many packages seemed to be getting updated for the latest version of Fedora, without retaining 100% compatibility with Fedora 9.  And while Fedora 9 is not particularly old in OS terms, it’s no longer a recent version – Fedora 11 is the current release, and an alpha of Fedora 12 is currently available for download.

So I decided now was a good time to upgrade to Fedora 11.  I considered Fedora 12 Alpha, but thought I’d best stick to an RTM release since I didn’t want to spend 4 hours a day every day for a month fixing it.

And so, what I learned was: back up your personal data.

After religiously backing up all the config files for everything I was running on the server, backing up all database, all websites, and making a list of everything installed, I went ahead and performed a clean install of Fedora 11.

And then I realised I had also been using Samba to provide file storage to our Windows machines.  Not only that, but I had previously decided to move (not copy, but move) all my photos from my desktop hard drive to the Linux server.  I have no idea why I didn’t notice them when I went through my list of things to back up, but miss them I did.  And now they are no more – apart from those in the gallery, every single photo I have ever taken was exploded into ones and zeroes for ever.

Much like being able to tell (but not necessarily care) when I’m boring someone with a particular conversation, I use website analytics to see where website visitors come from, what they look at, and how frequently they return. Yes, some people come back. In fact, most people do. And the most popular pages are actually the NSLU2 guides which describe how to configure a very small, low-power Linux server based on a Linksys NAS device.

I have no delusions about the overall importance of nikrivers.com; nobody knows about it, it doesn’t appear very high up in any Google results unless you search for my name, and even my mum keeps forgetting how to find it (though she also has to keep asking me how to find the Internet, because “it’s gone from my desktop”).

That said, 101 different people have visited my NSLU2 guides in the last 20 days alone, about a third of which are visitors who have never been to my website before. Even after Linksys has discontinued the NSLU2, it still remains popular as a web and mail server.

Linksys NSLU2

About the NSLU2

The Linksys NSLU2 is a lovely little NAS storage server with 2 USB ports and an Ethernet LAN port. You can attach two USB hard drives to it and connect it to your network, and use it as a file server.

Or you can download alternative firmware, run a customised version of Linux on it, and use it as a mail server, a torrent server, a print server, a web server, and more.

It supports NTFS, USB hubs, and flash drives. It’s silent, draws no more than 10W of power, runs at 266MHz, and is the size of two Weetabix.

And it’s known as a Slug.

Full details, including the data sheet, user guide and firmware, can be found over at the Linksys website, and some information is available at Wikipedia.

The NSLU2 is now discontinued, but don’t let that put you off – this will only really affect you if you need to return your Slug for replacement under warranty. Flashing your firmware voids the warranty (as does making hardware modifications, which a lot of users seem to do), so its discontinuation seems hardly relevant.

Opinion seems to be divided on this issue, with a few differing viewpoints given on Paul Hutchinson’s blog. Regardless, I predict the NSLU2 will maintain a cult following of users and developers for quite some time.

Where to buy

These things aren’t particularly easy to get hold of in the UK at a decent price, but LambdaTek has them available at around £70, although at dwindling stock levels.

Or, if you’re in the market for a second-hand Slug, I have a spare for which I cannot find a use; I’m more than happy to consider offers on it. It’s the 266MHz version (as opposed to the 133MHz version) and has a UK power supply. For use outside the UK, a travel adapter should be fine since the power supply is auto-switching, or a generic 5V-2A power supply would do the trick.

The Linksys NSLU2 is no longer available – it seems all retailers have run down their stocks.  You might be lucky on eBay, but if you don’t own one already I recommend you find an alternative.

Finding more help

The NSLU2 community is at www.nslu2-linux.org, where there’s a LOT of information. However, it’s a community built on voluntary user contribution, so the information is sometimes incomplete, outdated, or otherwise inaccurate. There’s also a lot of information about the available packages out there on the Internet, but it doesn’t necessarily relate to the NSLU2 and often assumes you will be using packages that simply aren’t available to you.

The Unslung firmware is the first custom firmware that most users try. The binary download, which also contains all the information you need to get started, can be found at www.slug-firmware.net.

Getting the Unslung firmware up and running on my Slug was relatively easy. However, finding the information I needed to configure certain software packages proved more difficult, even though the actual installation of each package was quite simple.

In order to document my Slug’s configuration and to provide a reference point to others, I have created step-by-step NSLU2 instructions for the following tasks:

• Enabling SSH access
• Setting up an email server using Postfix and Dovecot
• Setting up an NTP server using ntp
• Setting up a print server using CUPS
• Configuring SSL and TLS for Postfix and Dovecot